An End to Alzheimer's?
By Ken Garber [Technology Review]
March 2001
Researchers have found a molecule they believe is a key culprit in the development of
Alzheimer’s disease. Now the race is on to block the molecule—and stop the disease in its
tracks.
When molecular biologist Bob Vassar joined biotech giant Amgen in 1996, his mother was
sufffering from advanced-stage Alzheimer'sdisease. For years he'd taken care of her himself,
since his father had died young. "Once she became incontinent I couldn't keep her at home
any more," he recalls with more than a trace of guilt. In a nursing home, Vassar's mother slid
rapidly downhill. In 1999, at age 78, she died—17 years after her diagnosis of Alzheimer's,
which gradually but inexorably drains its victims of memory, judgment and reason.
Vassar was helpless to stop, or even slow, his mother's descent into dementia. But unlike
most Alzheimer's caregivers, he was in a position tohelp others avoid her fate. At Amgen,
in Thousand Oaks, CA, he designed and implemented an ingenious method for isolating
the gene for an enzyme called beta-secretase—a key culprit in the disease. "It was a
high-risk project, and there was no guarantee we could get it," says Vassar. In fact, his
group individually tested 860,000 gene copies before finding beta-secretase and publishing
the discovery in late 1999.
With success came hope. The U.S. Food and Drug Administration has approved three drugs
for Alzheimer's disease, but these only temporarily improve brain function, without slowing or
stopping progression of the disease. Other treatments are in advanced development (see
"Bulging Pipeline"), but none has yet shown good long-term results. The discovery of
beta-secretase, on the other hand, opens the possibility of halting the disease. "It's a huge leap
forward," says Lennart Mucke, director of the Gladstone Institute of Neurological Disease at the
University of California, San Francisco. Possession of beta-secretase (discovered, almost
simultaneously, by three other drug companies) has now set off a frenzied race to find and test
a drug to block the enzyme and stop Alzheimer's in its tracks.
Untangling Alzheimer's
A lot is at stake. Over four million people in the United States suffer from Alzheimer's, which
strikes at least 35 percent of those over 85 years old. According to gerontologist Mark Monane,
a stock analyst for Needham in New York, the direct and indirect costs of the disease are more
than $100 billion a year in this country alone. "There's a huge public outcry to have effective treatments available," Monane points out.
The financial stakes also are enormous. "By the year 2005 or 2010 the market could approach $8 billion," says Alzheimer's researcher Kevin Felsenstein of Bristol-Myers Squibb's pharmaceutical lab in Wallingford, CT. Any drug would likely be taken for life—a pharmaceutical company's dream. "If anything is going to define what a megablockbuster is, I think this would," says Felsenstein.
Yet, until very recently, Alzheimer's was so poorly understood that scientists despaired of
finding a treatment, much less a cure. "Fifteen years ago we had no clue about how to treat
this disease," said University of Pittsburgh researcher Steve DeKosky at the 2000 World
Alzheimer's Congress. That confusion has persisted since 1906, when German neurologist
Alois Alzheimer performed an autopsy on a woman in her 50s. In the woman's brain,
Alzheimer found "tangled bundles" inside neurons, and dense deposits of a "peculiar
substance" between the nerve cells. But what these "tangles" and "plaques" consisted of
remained a mystery for three-quarters of a century.
Then, in the mid-1980s, scientists determined that the tangles consisted of a protein called
tau, and the plaques contained a small protein fragment—or peptide—called beta-amyloid,
which was a piece of a larger "amyloid precursor protein." Which feature causes Alzheimer's
disease—tau or beta-amyloid (which is sometimes called BAP)—has been hotly debated ever
since, leading to acrimonious battles between so-called Tauists and BAPtists. The BAPtist
camp, led by Harvard neurologist Dennis Selkoe, gradually prevailed, although not all Tauists
have surrendered. "I would guess that probably 80 percent or more in the field would now
agree that amyloid plays a critical role," says Amgen molecular biologist Martin Citron, who
worked in Selkoe's lab from 1992 to 1997. Selkoe, Citron and others believe beta-amyloid is
either directly toxic to cells, or provokes inflammation that kills neurons, or both.
By the early 1990s, researchers had pieced together a model of what might be going on in
Alzheimer's patients' brains. Amyloid precursor protein grows like a blade of grass out of the
membrane that covers each nerve cell. Normally, researchers hypothesized, two scissors-like
enzymes called alpha-secretase and gamma-secretase work in series to clip the amyloid
precursor protein first near its base and then right at the base; the short peptide that results
from the second cut is harmless. Sometimes, though, rather than the alpha enzyme, a beta
version makes the first cut—and in a slightly different spot. When the gamma enzyme comes
along and makes the second cut, the resulting fragment is beta-amyloid. In Alzheimer's, the
theory goes, beta-amyloid piles up between the brain cells, forming plaques and gradually
killing surrounding cells.
Blunting the Scissors
Although the cutting enzymes themselves proved maddeningly elusive, this model of
Alzheimer's offered an obvious strategy for attacking the disease: block the enzymes and
prevent the plaques. Since 1992, many drug companies have been looking for "secretase
inhibitors," molecules that would block either gamma- or beta-secretase. One compound, a
gamma-secretase inhibitor discovered at Bristol-Myers Squibb using mass screening
techniques, entered early human trials in April 2000. "We're on the verge now of either
preventing amyloid deposits from building up, inhibiting the production of amyloid, or actually
being able to reverse plaque deposition," says Felsenstein, who leads Bristol-Myers Squibb's
amyloid program. Other pharmaceutical firms—including DuPont, Merck, Elan and Eli
Lilly—are testing gamma-secretase inhibitors as well, but haven't yet disclosed human trials.
But the drug-discovery process has been agonizingly slow. Just randomly sprinkling compounds
on cells and measuring amyloid levels requires both time and luck to get a good hit. In fact,
many drug companies ignored the secretases, since without the enzymes in hand, there's no
way to know how specifically a compound is targeting them—making toxic side effects largely
unpredictable.
Finding beta-secretase changes everything. Now medicinal chemists can design molecules to
fit precisely into the enzyme's "active site." In theory, such drugs should be exquisitely specific,
avoiding the worst side effects. Citron says Amgen is very excited about that possibility. The
company has figured out the three-dimensional structure of the beta enzyme and is fashioning
molecules to block its activity.
Industry insiders say that Amgen has lots of competition, though companies are more open
about their gamma programs. GlaxoSmithKline of London, another company to have isolated
the beta-secretase gene, is one player officially working to inhibit it. And Dublin, Ireland-based
Elan and its partner in Peapack, NJ, Pharmacia, have also launched a major push. "Both of us
view beta-secretase as a terrific target," says Dale Schenk, vice president of discovery research
for Elan. "I don't think it's going to be terribly long before the field has clinical candidates."
Schenk's optimism is based, in part, on an obvious precedent: AIDS. That's because
beta-secretase is a protease, or protein-cutting enzyme, in the same class as the HIV protease,
which proved to be a great drug target. Once the HIV-protease structure was discovered in
1989, it took less than three years to get "protease inhibitors" into the clinic. These drugs have
changed AIDS from a death sentence into a usually manageable condition. "Pharmaceutical
companies like sure things," says University of South Florida Alzheimer's researcher Huntington
Potter. Blocking enzymes "is something they can do easily and be sure that they have
something fairly successful at the end."
Progress Inhibited?
But there is a big question: what if the amyloid theory is wrong? If it is, secretase inhibitors
would be useless. "The field has been 'sold,' or has willingly bought into, the hypothesis that
amyloid deposits, or possibly the precursors to the amyloid deposits, actually cause the
disease," says Potter. "It's probably true." But other factors—tangle formation, for example, or
disruption of neurons' stores of the calcium ions critical to nerve firing—could prove to be the
real cause of Alzheimer's. In that case, "getting rid of [amyloid] might make a cleaner brain,
but it might not make a more functioning brain," says Potter. "We won't know until [drugs] are
tried."
Even if the amyloid hypothesis proves correct, side effects could ultimately doom secretase
inhibitors. Gamma-secretase, for example, is closely related to a protein that is responsible for
processing another protein called Notch. Since Notch is crucial for human development and
cell division, a drug that blocked gamma-secretase might wind up causing chemotherapy-like
side effects—something unacceptable in a drug that's taken for life. "The Notch thing has
everybody scared," says one drug-company scientist.
Bristol-Myers Squibb and others are undeterred. Felsenstein says his company's drug, in mice,
inhibits gamma-secretase approximately 20 times more than it interferes with Notch activity:
"That gave us a window of opportunity." Felsenstein hopes that blocking perhaps 40 percent of
the enzyme's activity will be enough to keep plaques from forming but not enough to cause
serious side effects. "[Notch] may be a theoretical issue, or it may be a show stopper," says
Amgen's Citron. "Nobody knows."
Beta-secretase inhibitors have their own question marks. Nobody knows what the enzyme's
normal function in the body is, so blocking it could conceivably cause anything from hair loss
to psychosis. And random screening of naturally occurring and synthetic compounds over the
last eight years turned up no inhibitors of the beta enzyme. "That's very strange," says
Felsenstein. "I don't know if that tells us that beta-secretase may not be a good target. Only
time will tell."
And if beta-secretase inhibitors work, companies will face another battle over the enzyme.
That's because, with four groups isolating the gene almost simultaneously, no one knows who
will have first rights to the molecule. GlaxoSmithKline may have the edge, because it applied
first for a patent—but before it knew the enzyme's function. "It's going to take an army of
lawyers to figure out who owns it," says Felsenstein. Meanwhile, the pressure to get to market
ahead of competitors is intense. As one drug-company researcher confided, "My CEO says it's
costing us $150,000 a day not to have an Alzheimer's drug on the market."
Stocking the Armamentarium
Though each company is desperate to be first, many will likely emerge with successful (and
lucrative) drugs. "The field is wide open," says Needham's Monane. "There's room for multiple
winners in the race." And, indeed, most researchers in the field agree that no single drug is
likely to work for every Alzheimer's patient.
Instead, it will probably take a "cocktail" approach, analogous to today's AIDS therapy. So,
although secretase inhibitors and a vaccine effort at Elan (see companion article "Injection of
Hope") hold the most promise now, other classes of drugs could also be useful. "We will
probably find out that many different things are contributing to this disease," says the University
of California, San Francisco's Mucke. "And it is worthwhile to develop treatments to hit each
one of those, and then see what combination will be most effective."
How soon will this happen? "I predict that we're going to have an armamentarium of drugs in
the next five years—10 years maximum—some or many of which will be effective," says Potter.
"It's going to be longer before we can say that we've cured the disease."
Meanwhile, Amgen's Bob Vassar continues working on beta-secretase. His family history of
Alzheimer's weighs on him constantly. He knows his mother may have had an inherited form of
the disease—one that could have been passed on to him. "I'm worried about myself, and
wondering what's going to happen to me," he says. Vassar, who's 44, thought of being tested
for the defective genes but decided against it, figuring it would only add to his stress. Instead,
he says, "I'm hoping in the back of my mind that either Amgen or another company comes up
with an effective inhibitor." He knows it won't happen overnight. "But hopefully, by the time I
need it, it will be there."
Ken Garber is a self-taught science journalist and avid rock climber from Michigan.
Companion Article
Injection of Hope
Dale Schenk's office at Elan's pharmaceutical division in South San Francisco is decorated
with framed clippings from Time, Newsweek and The New York Times, among others. The clips
commemorate a stunning finding that Elan publicized in 1999: injecting the beta-amyloid
peptide into mice that have an experimental form of Alzheimer's eliminated the amyloid
plaques that many believe are the cause of brain damage in the disease.
"It's almost an obvious experiment, but we all assumed it would be a bad thing to do rather
than a good thing," says University of South Florida Alzheimer's researcher Huntington
Potter. Only Schenk, a vice president for discovery research at Elan, envisioned that
beta-amyloid might work as a vaccine, provoking a mouse's own immune system to remove
the plaques. Schenk "gets sole personal credit," says Potter, who considers the vaccine
"the most encouraging potential therapy on the horizon." Elan and Madison, NJ-based American Home Products together launched a clinical trial of the vaccine in January 2000.
Schenk stresses that the vaccine work is hardly his first foray into Alzheimer's. "I feel,
sometimes, a bit like an actor," he says. "Once people associate you with something,
they forget what else you've done." But despite his accomplishments, Schenk faced rampant skepticism, even within Elan, when he proposed the vaccine. And Schenk himself was shocked
at his spectacular results. "I can't say that I expected it to work as well as it did," he recalls.
"I was very surprised."
Schenk says that the vaccine could be widely available in "three to six years." Based on animal
studies, he says, immunization promises to affect the disease more profoundly than secretase
inhibitors—which could reach the market sooner. But the vaccine faces multiple hurdles. On
the one hand, it might not provoke as robust an immune response in humans as it does in
mice. On the other, it might overstimulate the immune system, causing damage to other
important proteins in the body. And getting rid of amyloid might turn out to be harmful, since it
may serve some important positive function in the body. Finally, immunization might help
some patients but accelerate the disease in others.
"We're trying a zillion different things," says Schenk, to address the various risks. One potentially
safer approach: instead of injecting the peptide to force the body to make antibodies, just
inject the antibodies. Schenk says the antibody method clears plaques in animals. Ultimately,
though, he agrees the only proof would be patients' improvement. "Do I think [the vaccine] is
going to work?" he asks. "Absolutely."
Bulging Pipleine
So far, only Bristol-Myers Squibb is known to be testing a secretase inhibitor in humans, but
several other types of Alzheimer's drugs are already in clinical trials in the United States.
[Chart] - technologyreview.com |
